[Air-L] Talk: "Emergent Analogical Reasoning in Large Language Models"

Nicholas Proferes Nicholas.Proferes at asu.edu
Mon Feb 27 11:13:15 PST 2023


Hello all,


ASU's interdisciplinary data-science research group, B²C², will be hosting a public Zoom talk --this Thursday (3/2) at 1pm MST-- from Dr. Taylor Webb (UCLA) on emergent analogical reasoning in large language models. The event is free, though registration is required.

When: Thursday, March 2nd at 1PM MST (20:00 UTC)

Location: Zoom

Registration Link: https://asu.zoom.us/meeting/register/tZYrcu-qrDIoH9wsqUB2CW-imRZldf5qIznP

Who: Dr. Taylor Webb, UCLA

Title: Emergent Analogical Reasoning in Large Language Models

Abstract: The recent advent of large language models — large neural networks trained on a simple predictive objective over a massive corpus of natural language — has reinvigorated debate over whether human cognitive capacities might emerge in such generic models given sufficient training data. Of particular interest is the ability of these models to reason about novel problems zero-shot, without any direct training on those problems. In human cognition, this capacity is closely tied to an ability to reason by analogy. Here, we performed a direct comparison between human reasoners and a large language model (GPT-3) on a range of analogical tasks, including a novel text-based matrix reasoning task closely modeled on Raven's Progressive Matrices. We found that GPT-3 displayed a surprisingly strong capacity for abstract pattern induction, matching or even surpassing human capabilities in most settings. Our results indicate that large language models such as GPT-3 have acquired an emergent ability to find zero-shot solutions to a broad range of analogy problems.

Bio: Dr. Taylor Webb is a postdoctoral scholar in the UCLA Department of Psychology, working with Keith Holyoak, Hongjing Lu, and Hakwan Lau. His research is situated at the interface between cognitive science and AI, with a particular emphasis on using neural network techniques to build cognitive models that are grounded in real-world perceptual inputs.

Looking forward to seeing you there,
Nicholas



Nicholas Proferes, PhD

Assistant Professor

School of Social and Behavioral Sciences

New College of Interdisciplinary Arts and Sciences

Arizona State University







More information about the Air-L mailing list